Starlink satellites can disturb observation even of those telescopes protected by radio-quiet zones.

  • tal@kbin.social
    link
    fedilink
    arrow-up
    9
    ·
    edit-2
    1 year ago

    Probably going to have to move to radio telescopes on the dark side of the moon or something. I mean, I seriously doubt that terrestrial users are going to let frequency go unused.

    For some users, maybe we could switch to lasers, which are more-directional – like, a hypothetical Laser Starlink would have one or a handful of lasers on a station that physically track a satellite or satellites. Problem is that that doesn’t work well with clouds – visible light is obstructed by them.

    Maybe it’s possible to use masers, but I assume that if it were technically easy and cheap, it would have been done by now.

      • tal@kbin.social
        link
        fedilink
        arrow-up
        6
        ·
        edit-2
        1 year ago

        I mean, the opportunity cost of not being able to use part of the frequency spectrum is also pretty big. And some of the structural elements are there to stand up to terrestrial conditions, like precipitation, wind, and much-stronger gravity. They wouldn’t need those on the Moon.

        I think a more-fundamental issue is that it imposes constraints on the direction in which one can be pulling data from. No great fix for that.

        EDIT: If this NASA project makes it to deployment, then there will be at least one up there.

        https://en.wikipedia.org/wiki/Lunar_Crater_Radio_Telescope

        If completed, the telescope would have a structural diameter of 1.3 km, and the reflector would be 350m in diameter.[3][4][5] Robotic lift wires and an anchoring system would enable origami deployment of the parabolic reflector.[6]

        • rankshank@kbin.social
          link
          fedilink
          arrow-up
          4
          ·
          1 year ago

          Don’t overlook the changes required for electronics that are able to operate in space. Since there’d be no atmospheric sheilding from radiation, the amount of additional silicon for error correction used per unit of compute is much higher. The capacity for cooling is also much lower on the moon, you’d essentially have to slap huge heatsinks on every component since you basically rely on radiation for heat dissipation. You’ll also constantly be fighting with the fact that every electrically conductive trace serves as an antenna, so the trace length vs component density for heat dissipation is going to be a constant battle. Then there is the limited availability of power.

          It all adds up to an entirely different class of device being able to be deployed in space. On earth we can just chuck high precision components around, throw swathes of power and cooling at it and call it a day. Rain and weather are a footnote compared to the design challenges space deployments represent.

        • AshDene@kbin.social
          link
          fedilink
          arrow-up
          4
          ·
          1 year ago

          Both “on the earth” and “on the moon” provide about the viewing angle of the sky (a semi-sphere). Unless we’re tracking an object with multiple of these spaced around the earth to get 24/7 recordings the moon doesn’t seem worse…

          Even then, with two of these you could put them opposite eachother just barely into the “dark side” (side facing away from earth) of the moon and get nearly 360 degree coverage. You’d have to not literally be on the boundary/leave an earth sized gap in the coverage, but it would be pretty damn close.

            • AshDene@kbin.social
              link
              fedilink
              arrow-up
              3
              ·
              1 year ago

              Satellites go beyond the moon, but not starlink satellites (or any future competing large mesh network of satellites), they are in in low orbit to minimize latency. I haven’t double checked with math or anything but I don’t think they should be high enough to be in sight of much more of the moon than the earth is.

      • FaceDeer@kbin.social
        link
        fedilink
        arrow-up
        2
        ·
        1 year ago

        The tradeoff being done here that makes me really excited for the future of astronomy is that Starlink is funding the development of Starship, which will in turn make space-based telescopy a hugely easier thing to do. So I’d gladly hand off a bit of spectrum pollution here on Earth (which comes with vastly improved global internet access) for Starship’s launch capacity.

        • lol3droflxp@kbin.social
          link
          fedilink
          arrow-up
          6
          ·
          1 year ago

          So being dependent upon the company that ruins the sky on earth but offers to get your science off planet (if starship will even work as promised in the end) is a good thing?

          • FaceDeer@kbin.social
            link
            fedilink
            arrow-up
            1
            ·
            1 year ago

            If Starship doesn’t work as promised then there will be no Starlink constellation in the long run. The two projects are dependent on each other. Starlink V2 satellites are necessary for the long-term profitability of the constellation, and Starlink V2 satellites can only be launched by Starship.

            The “dependency” is only a “dependency” in the sense that SpaceX Starship will be insanely cheap to use compared to any existing competitor. Maybe some of those other well-established space launch companies should have been working on making their launchers better too. I’m sure they’ll be scrambling to do so now that they face actual competition.

            • lol3droflxp@kbin.social
              link
              fedilink
              arrow-up
              1
              ·
              1 year ago

              Maybe, however last time I checked starship still had significant issues that have some chance of not getting resolved and flacon 9 launches are still quite expensive but that may have changed since then

              • FaceDeer@kbin.social
                link
                fedilink
                arrow-up
                1
                ·
                1 year ago

                SpaceX is a for-profit company, so you can expect them to price their launches only a little bit lower than their competitors even if the cost of the launch is dramatically lower. That gives them the most profit. If you want the price to go down significantly then you’ll need to find someone else who can start actually reusing their rockets to get their costs into the same ballpark as SpaceX.

                What specific significant issues did you hear that Starship had? NASA is confident enough in their chances that the success of the Artemis program was literally dependent on Starship being successful (the human lander is a modified Starship), and the design has changed a lot even since their previous test launch.

                • NotMyOldRedditName@kbin.social
                  link
                  fedilink
                  arrow-up
                  2
                  ·
                  edit-2
                  1 year ago

                  I don’t think starship is going to be priced like that. They’ve long been saying it’s going to dramatically reduce cost to orbit for everyone.

                  Will they make it more expensive than what it would cost them for a starlink v2 launch, sure, but it’s not gonna be priced per kg just below the next cheapest non resuseable rocket either.

                • lol3droflxp@kbin.social
                  link
                  fedilink
                  arrow-up
                  1
                  ·
                  1 year ago

                  So far the booster doesn’t seem to work since using so many engines hasn’t been solved yet. Maybe they will figure it out.

                  • FaceDeer@kbin.social
                    link
                    fedilink
                    arrow-up
                    1
                    ·
                    1 year ago

                    The booster they test-launched used a hodgepodge of every engine that they’d built so far, with designs that were already obsolete when it was launched. It’s hardly surprising that some of them failed. The next test launch will be using a set of mass-produced engines with a more refined and consistent design, and they’re entirely removing the hydraulic system that was the cause of the first test launch’s failure.

                    Again, NASA is confident this will work out. The HLS contract is in the critical path for the Artemis Moon landing and they gave it to SpaceX in a form that depends on Superheavy working. This isn’t just my opinion here.